Assessing Transformation Pathways - IPCC
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
6 Assessing Transformation Pathways Coordinating Lead Authors: Leon Clarke (USA), Kejun Jiang (China) Lead Authors: Keigo Akimoto (Japan), Mustafa Babiker (Sudan / Saudi Arabia), Geoffrey Blanford (USA / Germany), Karen Fisher-Vanden (USA), Jean-Charles Hourcade (France), Volker Krey (IIASA / Germany), Elmar Kriegler (Germany), Andreas Löschel (Germany), David McCollum (IIASA / USA), Sergey Paltsev (Belarus / USA), Steven Rose (USA), Priyadarshi R. Shukla (India), Massimo Tavoni (Italy), Bob van der Zwaan (Netherlands), Detlef P. van Vuuren (Netherlands) Contributing Authors: Hannes Böttcher (Austria / Germany), Katherine Calvin (USA), Katie Daenzer (USA), Michel den Elzen (Netherlands), Subash Dhar (India / Denmark), Jiyong Eom (Republic of Korea), Samuel Hoeller (Germany), Niklas Höhne (Germany), Nathan Hultman (USA), Peter Irvine (UK / Germany), Jessica Jewell (IIASA / USA), Nils Johnson (IIASA / USA), Amit Kanudia (India), Agnes Kelemen (Hungary), Klaus Keller (Germany / USA), Peter Kolp (IIASA / Austria), Mark Lawrence (USA / Germany), Thomas Longden (Australia / Italy), Jason Lowe (UK), André Frossard Pereira de Lucena (Brazil), Gunnar Luderer (Germany), Giacomo Marangoni (Italy), Nigel Moore (Canada / Germany), Ionna Mouratiadou (Greece / Germany), Nils Petermann (Germany), Philip Rasch (USA), Keywan Riahi (IIASA / Austria), Joeri Rogelj (Switzerland / Belgium), Michiel Schaeffer (Netherlands / USA), Stefan Schäfer (Germany), Jan Sedlacek (Switzerland), Laura Sokka (Finland), Christoph von Stechow (Germany), Ian Sue Wing (Trinidad and Tobago / USA), Naomi Vaughan (UK), Thilo Wiertz (Germany), Timm Zwickel (Germany) Review Editors: Wenying Chen (China), John Weyant (USA) Chapter Science Assistant: Laura Sokka (Finland) 413
Assessing Transformation Pathways Chapter 6 This chapter should be cited as: Clarke L., K. Jiang, K. Akimoto, M. Babiker, G. Blanford, K. Fisher-Vanden, J.-C. Hourcade, V. Krey, E. Kriegler, A. Löschel, D. McCollum, S. Paltsev, S. Rose, P. R. Shukla, M. Tavoni, B. C. C. van der Zwaan, and D.P. van Vuuren, 2014: Assessing Transformation Pathways. In: Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Edenhofer, O., R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J. Savolainen, S. 6 Schlömer, C. von Stechow, T. Zwickel and J.C. Minx (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. 414
Chapter 6 Assessing Transformation Pathways Contents Executive Summary������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 418 6 6.1 Introduction ���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 420 6.1.1 Framing and evaluating transformation pathways ���������������������������������������������������������������������������������������������������������������� 420 6.1.2 New mitigation scenarios since AR4������������������������������������������������������������������������������������������������������������������������������������������������ 420 6.1.2.1 Non-idealized international implementation scenarios���������������������������������������������������������������������������������������������� 421 6.1.2.2 Limited technology scenarios�������������������������������������������������������������������������������������������������������������������������������������������������� 421 6.2 Tools of analysis������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 422 6.2.1 Overview of integrated modelling tools���������������������������������������������������������������������������������������������������������������������������������������� 422 6.2.2 Overview of the scenario ensemble for this assessment������������������������������������������������������������������������������������������������������ 423 6.2.3 Uncertainty and the interpretation of large scenario ensembles ���������������������������������������������������������������������������������� 423 6.2.4 Interpretation of model inability to produce particular scenarios �������������������������������������������������������������������������������� 424 6.3 Climate stabilization: Concepts, costs and implications for the macro economy, sectors and technology portfolios, taking into account d ifferences across regions�������������������������� 424 6.3.1 Baseline scenarios�������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 424 6.3.1.1 Introduction to baseline scenarios���������������������������������������������������������������������������������������������������������������������������������������� 424 6.3.1.2 The drivers of baseline energy-related emissions���������������������������������������������������������������������������������������������������������� 424 6.3.1.3 Baseline emissions projections from fossil fuels and industry�������������������������������������������������������������������������������� 425 6.3.1.4 Baseline CO2 emissions from land use and emissions of non-CO2 gases���������������������������������������������������������� 426 6.3.1.5 Baseline radiative forcing and cumulative carbon emissions���������������������������������������������������������������������������������� 428 6.3.2 Emissions trajectories, concentrations, and temperature in transformation pathways �������������������������������������� 428 6.3.2.1 Linking between different types of scenarios������������������������������������������������������������������������������������������������������������������ 428 6.3.2.2 The timing of emissions reductions: The influence of technology, policy, and overshoot ������������������������ 433 6.3.2.3 Regional roles in emissions reductions ������������������������������������������������������������������������������������������������������������������������������ 434 6.3.2.4 Projected CO2 emissions from land use ���������������������������������������������������������������������������������������������������������������������������� 435 6.3.2.5 Projected emissions of other radiatively important substances ���������������������������������������������������������������������������� 436 6.3.2.6 The link between concentrations, radiative forcing, and temperature �������������������������������������������������������������� 438 6.3.3 Treatment of impacts and adaptation in transformation pathways ������������������������������������������������������������������������������ 441 6.3.4 Energy sector in transformation pathways ���������������������������������������������������������������������������������������������������������������������������������� 443 6.3.5 Land and bioenergy in transformation pathways �������������������������������������������������������������������������������������������������������������������� 445 415
Assessing Transformation Pathways Chapter 6 6.3.6 The aggregate economic implications of transformation pathways������������������������������������������������������������������������������ 448 6.3.6.1 Overview of the aggregate economic implications of mitigation ������������������������������������������������������������������������ 448 6.3.6.2 Global aggregate costs of mitigation in idealized implementation scenarios������������������������������������������������ 449 6.3.6.3 The implications of technology portfolios for aggregate global economic costs������������������������������������������ 453 6.3.6.4 Economic implications of non-idealized international mitigation policy implementation������������������������ 459 6.3.6.5 The interactions between policy tools and their implementation, pre-existing taxes, 6 market failures, and other distortions��������������������������������������������������������������������������������������������������������������������������������� 455 6.3.6.6 Regional mitigation costs and effort-sharing regimes ���������������������������������������������������������������������������������������������� 456 6.4 Integrating long- and short-term perspectives�������������������������������������������������������������������������������������������������������������������������� 462 6.4.1 Near-term actions in a long-term perspective���������������������������������������������������������������������������������������������������������������������������� 462 6.4.2 Near-term emissions and long-term transformation pathways ���������������������������������������������������������������������������������������� 462 6.4.3 evelopment of institutional capacity�������� 464 The importance of near-term technological investments and d 6.5 Integrating technological and societal change�������������������������������������������������������������������������������������������������������������������������� 466 6.5.1 Technological change�������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 466 6.5.2 Integrating societal change ������������������������������������������������������������������������������������������������������������������������������������������������������������������ 467 6.6 Sustainable development and transformation pathways, taking into account differences across regions���������������������������������������������������������������������������������������������������������������������������������������������������������� 468 6.6.1 Co-benefits and adverse side-effects of mitigation measures: Synthesis of sectoral information and linkages to transformation pathways �������������������������������������������������������� 472 6.6.2 Transformation pathways studies with links to other policy objectives���������������������������������������������������������������������� 472 6.6.2.1 Air pollution and health�������������������������������������������������������������������������������������������������������������������������������������������������������������� 473 6.6.2.2 Energy security�������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 475 6.6.2.3 Energy access ���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 476 6.6.2.4 Employment�������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 476 6.6.2.5 Biodiversity conservation ���������������������������������������������������������������������������������������������������������������������������������������������������������� 476 6.6.2.6 Water use ������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 477 6.6.2.7 Integrated studies of multiple objectives�������������������������������������������������������������������������������������������������������������������������� 477 6.7 Risks of t ransformation pathways������������������������������������������������������������������������������������������������������������������������������������������������������������ 478 6.8 Integrating sector analyses and t ransformation scenarios�������������������������������������������������������������������������������������������� 478 6.8.1 The sectoral composition of GHG emissions along transformation pathways���������������������������������������������������������� 478 6.8.2 Mitigation from a cross-sectoral perspective: Insights from integrated models ���������������������������������������������������� 479 6.8.3 Decarbonizing energy supply��������������������������������������������������������������������������������������������������������������������������������������������������������������� 480 416
Chapter 6 Assessing Transformation Pathways 6.8.4 Energy demand reductions and fuel switching in end-use sectors���������������������������������������������������������������������������������� 480 6.8.5 Options for bioenergy production, reducing land-use change emissions, and creating land-use GHG sinks�������������������������������������������������������������������������������������������������������������������������������������������������������� 484 6.9 Carbon and radiation management and other geo-engineering options 6 including environmental risks������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 484 6.9.1 Carbon dioxide removal�������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 485 6.9.1.1 Proposed carbon dioxide removal methods and characteristics���������������������������������������������������������������������������� 485 6.9.1.2 Role of carbon dioxide removal in the context of transformation pathways�������������������������������������������������� 486 6.9.2 Solar radiation management���������������������������������������������������������������������������������������������������������������������������������������������������������������� 486 6.9.2.1 Proposed solar radiation management methods and characteristics ���������������������������������������������������������������� 486 6.9.2.2 The relation of solar radiation management to climate policy and transformation pathways�������������� 487 6.9.3 Summary���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 489 6.10 Gaps in knowledge and data�������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 489 6.11 Frequently Asked Questions���������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 490 References ���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 491 417
Assessing Transformation Pathways Chapter 6 Executive Summary Scenarios with more overshoot exhibit less mitigation today, but they often rest on the assumption that future decision makers deploy CDR technologies at large scale. An assessment in this chapter of geophysi- Stabilizing greenhouse gas (GHG) concentrations will require large- cal climate uncertainties consistent with the dynamics of Earth System scale transformations in human societies, from the way that we pro- Models assessed in Working Group I (WG I) provides estimates of the duce and consume energy to how we use the land surface. A natural temperature implications of different emissions pathways. This assess- 6 question in this context is what will be the ‘transformation pathway’ ment found that the likelihood of exceeding temperature goals this towards stabilization; that is, how do we get from here to there? The century increases with peak concentration levels, which are higher in topic of this chapter is transformation pathways. The chapter is pri- overshoot scenarios. [6.3.2] marily motivated by three questions. First, what are the near-term and future choices that define transformation pathways, including the goal All major-emitting regions make substantial reductions from itself, the emissions pathway to the goal, technologies used for and their baseline CO2eq emissions over the century in scenarios sectors contributing to mitigation, the nature of international coordi- that bring atmospheric concentrations to about 550 ppm CO2eq nation, and mitigation policies? Second, what are the key characteris- or below by 2100 (high confidence). In most scenarios collected for tics of different transformation pathways, including the rates of emis- this assessment that reach concentrations of about 550 ppm CO2eq by sions reductions and deployment of low-carbon energy, the magnitude 2100, global CO2eq emissions are reduced by more than 50 %, and in and timing of aggregate economic costs, and the implications for other some cases by more than 100 %, by the end of the century relative to policy objectives such as those generally associated with sustainable 2010 levels. The CO2eq emissions are brought to near or below zero development? Third, how will actions taken today influence the options by 2100 in the majority of the scenarios reaching concentrations of that might be available in the future? As part of the assessment in this about 450 ppm CO2eq by 2100. In large part because baseline emis- chapter, data from over 1000 new scenarios published since the IPCC sions from the countries not part of the Organisation for Economic Co- Fourth Assessment Report (AR4) were collected from integrated mod- operation and Development (OECD) in 1990 are projected to outstrip elling research groups, many from large-scale model intercomparison those from the OECD-1990 countries, the total CO2eq reductions from studies. In comparison to AR4, new scenarios, both in this AR5 dataset baseline occurring in the non-OECD-1990 countries are larger than in and more broadly in the literature assessed in this chapter, consider the OECD-1990 countries, particularly in scenarios that cost-effectively more ambitious concentration goals, a wider range of assumptions allocate emissions reductions across countries. Emissions peak earlier about technology, and more possibilities for delays in additional global in the OECD-1990 countries than in the non-OECD-1990 countries in mitigation beyond that of today and fragmented international action. these cost-effective scenarios. [6.3.2] Atmospheric concentrations in baseline scenarios collected for Bringing concentrations to about 550 ppm CO2eq or below by this assessment (scenarios without additional efforts to con- 2100 will require large-scale changes to global and national strain emissions beyond those of today) all exceed 450 parts energy systems, and potentially to the use of land; these per million (ppm) carbon dioxide-equivalent (CO2eq) by 2030 changes are inconsistent with both long- and short-term trends and lie above the RCP 6.0 representative concentration path- (high confidence). Accelerated electrification of energy end use, cou- way in 2100 (770 ppm CO2eq in 2100); the majority lie below pled with decarbonization of the majority of electricity generation by the RCP 8.5 concentration pathway in 2100 (1330 ppm CO2eq 2050 and an associated phaseout of freely emitting coal generation, in 2100) (high confidence). The scenario literature does not system- is a common feature of scenarios reaching about 550 ppm CO2eq or atically explore the full range of uncertainty surrounding development less by 2100. Scenarios suggest that sectors currently using liquid fuel pathways and the possible evolution of key drivers such as popula- are more costly to decarbonize than electricity and may be among the tion, technology, and resources. However, the baseline scenarios do last sectors to be decarbonized for deep CO2 emissions reductions. nonetheless strongly suggest that absent explicit efforts at mitigation, Scenarios articulate very different changes in the land surface, reflect- cumulative CO2 emissions since 2010 will exceed 700 GtCO2 by 2030, ing different assumptions about the potential for bioenergy produc- exceed 1500 GtCO2 by 2050, and potentially be well over 4000 GtCO2 tion, afforestation, and reduced deforestation. Studies indicate a large by 2100. [Section 6.3.1] potential for energy use reductions, but also demonstrate that these reductions will not be sufficient by themselves to constrain GHG emis- Scenarios can be distinguished by the long-term concentration sions. [6.3.4, 6.3.5, 6.8] level they reach by 2100; however, the degree to which concen- trations exceed (overshoot) this level before 2100 is also impor- Estimates of the aggregate economic costs of mitigation vary tant (high confidence). The large majority of scenarios produced in the widely, but increase with stringency of mitigation (high confi- literature that reach about 450 ppm CO2eq by 2100 are characterized dence). Most scenario studies collected for this assessment that are by concentration overshoot facilitated by the deployment of carbon based on the idealized assumptions that all countries of the world begin dioxide removal (CDR) technologies. Many scenarios have been con- mitigation immediately, there is a single global carbon price applied to structed to reach about 550 ppm CO2eq by 2100 without overshoot. well-functioning markets, and key technologies are available, estimate 418
Chapter 6 Assessing Transformation Pathways that reaching about 450 ppm CO2eq by 2100 would entail global con- implications for the challenge of achieving concentration goals sumption losses of 1 – 4 % in 2030 (median of 1.7 %), 2 – 6 % in 2050 (high confidence). Many models in recent multi-model comparisons (median of 3.4 %), and 3 – 11 % in 2100 (median of 4.8 %) relative to could not produce scenarios reaching approximately 450 ppm CO2eq what would happen without mitigation. These consumption losses cor- by 2100 with broadly pessimistic assumptions about key mitigation respond to an annual average reduction of consumption growth of 0.06 technologies. Large-scale deployment of CDR technologies in particular to 0.20 percentage points from 2010 to 2030 (median of 0.09), 0.06 is relied upon in many of these scenarios in the second-half of the cen- to 0.17 percentage points through 2050 (median of 0.09), and 0.04 to tury. For those models that could produce such scenarios, pessimistic 6 0.14 percentage points over the century (median of 0.06). To put these assumptions about important technologies for decarbonizing non-elec- losses in context, studies assume annual average consumption growth tric energy supply significantly increased the discounted global mitiga- rates without mitigation between 1.9 % and 3.8 % per year until 2050 tion costs of reaching about 450 ppm and about 550 ppm CO2eq by and between 1.6 % and 3.0 % per year over the century. These growth the end of the century, with the effect being larger for more stringent rates correspond to increases in total consumption from roughly four- goals. These studies also showed that reducing energy demand can fold to over ten-fold over the century. Costs for maintaining concen- potentially decrease mitigation costs significantly. [6.3.2, 6.3.4, 6.3.6, trations at around 550 ppm CO2eq are estimated to be roughly one- 6.4] third to two-thirds lower. Substantially higher and lower cost estimates have been obtained based on assumptions about less idealized policy Mitigation efforts will influence the costs of meeting other implementations, interactions with pre-existing distortions, non-climate policy objectives. Recent studies indicate that climate policies market failures, or complementary policies. (Limits on technology and significantly reduce the costs of reaching energy security and delayed mitigation are discussed below.) [6.3.6] air quality objectives (medium evidence, high agreement). The asso- ciated economic implications for these objectives are not taken into Effort-sharing frameworks could help address distributional account in most scenario studies. Sectoral studies suggest that the issues and decouple regional mitigation investments from potential for co-benefits of energy end-use mitigation measures out- financial burdens, but could be associated with significant inter- weighs the potential for adverse side-effects, whereas the evidence national financial flows (medium confidence). In the absence of suggests this may not be the case for all supply-side and AFOLU mea- effort-sharing frameworks, cost-effectively allocating emissions across sures. The overall welfare implications associated with these additional countries would yield an uneven distribution of mitigation costs. Sce- objectives have not been assessed thoroughly in the literature. [6.6] narios indicate that this would lead to higher relative costs in develop- ing economies as well as for many fossil fuel exporters. Studies explor- There is uncertainty about the potential of geoengineering by ing effort-sharing frameworks in the context of a global carbon market CDR or solar radiation management (SRM) to counteract climate estimate that the financial flows to ameliorate this asymmetry could change, and all techniques carry risks and uncertainties (high be on the order of hundreds of billions of USD per year before mid-cen- confidence). A range of different SRM and CDR techniques has been tury to bring concentrations to about 450 ppm CO2eq in 2100. [6.3.6] proposed, but no currently existing technique could fully replace miti- gation or adaptation efforts. Nevertheless, many low-GHG concentra- Emissions through 2030 will have strong implications for the tion scenarios rely on two CDR techniques, afforestation and biomass challenges of, and options for, bringing concentrations to about energy with carbon dioxide capture and storage (BECCS), which some 450 to about 500 ppm CO2eq by the end of the twenty-first cen- studies consider to be comparable with conventional mitigation meth- tury (high confidence). The vast majority of cost-effective scenarios ods. Solar radiation management could reduce global mean tempera- leading to 2100 concentrations of about 450 to about 500 ppm CO2eq tures, but with uneven regional effects, for example on temperature and are characterized by 2030 emissions roughly between 30 GtCO2eq precipitation, and it would not address all of the impacts of increased and 50 GtCO2eq. Scenarios with emissions above 55 GtCO2eq in 2030 CO2 concentrations, such as ocean acidification. Techniques requiring are predominantly driven by delays in additional mitigation relative large-scale interventions in the earth system, such as ocean fertilization to what would be most cost-effective. These scenarios are character- or stratospheric aerosol injections, carry significant risks. Although pro- ized by substantially higher rates of emissions reductions from 2030 posed geoengineering techniques differ substantially from each other, to 2050, a larger reliance on CDR technologies in the long term, and all raise complex questions about costs, risks, governance, and ethical higher transitional and long-term economic impacts. Due to these implications of research and potential implementation. [6.9] challenges, many models with 2030 emissions in this range could not produce scenarios reaching about 450 ppm CO2eq in 2100. Studies Despite the advances in our understanding of transformation path- confirm that delaying additional mitigation through 2030 has substan- ways since AR4, many avenues of inquiry remain unanswered. Impor- tially larger influence on the subsequent challenges of mitigation than tant future research directions include the following: development of delaying only through 2020. [6.3.2, 6.4] a broader set of socioeconomic and technological storylines to sup- port development of scenarios; scenarios explicitly pursuing a wider The availability of key technologies and improvements in the set of climate goals, including those related to temperature change; cost and performance of these technologies will have important more mitigation scenarios that include impacts from, and adaptations 419
Assessing Transformation Pathways Chapter 6 to, a changing climate, including energy and land use systems critical The second concept is that transformation pathways can be distin- for mitigation; expanded treatment of the benefits and risks of CDR guished from one another in important ways. Weighing the character- and SRM options; expanded treatment of co-benefits and adverse istics of different pathways is the way in which deliberative decisions side-effects of mitigation pathways; improvements in the treatment about transformation pathways would be made. Although measures of and understanding of mitigation options and responses in end-use sec- aggregate economic implications have often been put forward as key tors in transformation pathways; and more sophisticated treatments deliberative decision making factors, these are far from the only char- 6 of land use and land use-based mitigation options in mitigation sce- acteristics that matter for making good decisions. Transformation path- narios. [6.10] ways inherently involve a range of tradeoffs that link to other national and policy objectives such as energy and food security, the distribu- tion of economic costs, local air pollution, other environmental factors associated with different technology solutions (e. g., nuclear power, 6.1 Introduction coal-fired carbon dioxide capture and storage (CCS)), and economic competitiveness. Many of these fall under the umbrella of sustainable development. 6.1.1 Framing and evaluating transformation A question that is often raised about particular stabilization goals pathways and transformation pathways to those goals is whether the goals or pathways are ‘feasible’. In many circumstances, there are clear physi- Stabilizing greenhouse gas (GHG) concentrations at any level will cal constraints that can render particular long-term goals physically require deep reductions in GHG emissions. Net global CO2 emissions, impossible. For example, if additinional mitigation beyond that of in particular, must eventually be brought to or below zero. Emissions today is delayed to a large enough degree and carbon dioxide removal reductions of this magnitude will require large-scale transformations in (CDR) options are not available (see Section 6.9), a goal of reaching human societies, from the way that we produce and consume energy 450 ppm CO2eq by the end of the 21st century can be physically impos- to how we use the land surface. The more ambitious the stabilization sible. However, in many cases, statements about feasibility are bound goal, the more rapid this transformation must occur. A natural question up in subjective assessments of the degree to which other character- in this context is what will be the transformation pathway toward sta- istics of particular transformation pathways might influence the ability bilization; that is, how do we get from here to there? or desire of human societies to follow them. Important characteristics include economic implications, social acceptance of new technolo- The topic of this chapter is transformation pathways. The chapter is gies that underpin particular transformation pathways, the rapidity motivated primarily by three questions. First, what are the near-term at which social and technological systems would need to change to and future choices that define transformation pathways including, for follow particular pathways, political feasibility, and linkages to other example, the goal itself, the emissions pathway to the goal, the tech- national objectives. A primary goal of this chapter is to illuminate these nologies used for and sectors contributing to mitigation, the nature characteristics of transformation pathways. of international coordination, and mitigation policies? Second, what are the key decision making outcomes of different transformation pathways, including the magnitude and international distribution of 6.1.2 New mitigation scenarios since economic costs and the implications for other policy objectives such AR4 as those associated with sustainable development? Third, how will actions taken today influence the options that might be available in Since the IPCC Fourth Assessment Report (AR4), the integrated mod- the future? elling community has produced a range of new transformation path- way scenarios. Major advances include an increase in the number of Two concepts are particularly important for framing any answers to scenarios exploring the following: low-concentration goals such as these questions. The first is that there is no single pathway to stabiliza- 450 ppm CO2eq; overshoot emissions trajectories with and without tion of GHG concentrations at any level. Instead, the literature eluci- CDR technologies; a variety of international mitigation policy configu- dates a wide range of transformation pathways. Choices will govern rations, including fragmented action and delays in additional mitiga- which pathway is followed. These choices include, among other things, tion beyond that of today; and the implications of variations in tech- the long-term stabilization goal, the emissions pathway to meet that nology cost, performance, and availability. The literature also includes goal, the degree to which concentrations might temporarily overshoot a small but growing set of scenarios and research exploring the link- the goal, the technologies that will be deployed to reduce emissions, age between mitigation and other policy objectives, an increasingly the degree to which mitigation is coordinated across countries, the sophisticated treatment of the role of land use in mitigation, and sce- policy approaches used to achieve these goals within and across coun- narios exploring non-market approaches to mitigation. Two particularly tries, the treatment of land use, and the manner in which mitigation is important categories for the discussion in this chapter are non-ideal- meshed with other policy objectives such as sustainable development. ized international implementation scenarios and scenarios with limits 420
Chapter 6 Assessing Transformation Pathways on technology cost, performance, or availability. These categories of is inconsistent with — typically less than — what would be called for to scenarios are discussed in more detail below. minimize the discounted, century-long costs of meeting a long-term goal such as 450 ppm CO2eq by 2100. These scenarios are intended to capture the implications of ‘delayed action’ or ‘delayed mitigation’ or 6.1.2.1 Non-idealized international implementation ‘constrained near-term ambition’. Mitigation is not undertaken ‘when’ scenarios it would be least expensive. The other set of scenarios includes those in which the price on carbon is not consistent across countries. Some 6 At the time of AR4, the majority of mitigation scenarios were based on countries reduce emissions more aggressively than others, particularly the idealized assumption that mitigation is undertaken where and in the near-term, so that mitigation is not undertaken ‘where’ it is least when it is least expensive. Such ‘idealized implementation’ scenarios expensive. These scenarios are intended to capture the implications assume the imposition of a global price on carbon that reaches across of ‘fragmented action’ or ‘delayed participation’. Non-idealized inter- countries, permeates all economic sectors within countries, and rises national implementation scenarios may include one or both of these over time in a way that will minimize discounted economic costs over deviations. a long period of time, typically through 2100. These are often referred to as ‘cost-effective’ scenarios, because they lead to the lowest aggre- gate global mitigation costs under idealized assumptions about the 6.1.2.2 Limited technology scenarios functioning of markets and economies (see Section 6.3.6). However, the reality of international strategies for mitigation is one of different Scenario research prior to AR4 emphasized the importance of tech- countries taking on mitigation at different times and using different nology in constraining the costs of mitigation. A range of individual and independent implementation approaches. Responding to this real- papers had made initial explorations of this space for more than a ity, the research community has produced a large set of ‘non-idealized’ decade before AR4. Since AR4, however, a range of new studies have international implementation scenarios for reaching long-term concen- emerged including large model intercomparison studies, that have tration goals. Often, but not always, non-idealized implementation is focused on the implications of limitations on technology cost, per- focused on the coming decades, with a transition toward idealized formance, availability on the cost and other characteristics of meet- implementation in the long run. In addition to individual papers (for ing concentration goals such as 450 ppm CO2eq by 2100. The large example, Richels et al., 2007; Edmonds et al., 2008; Luderer et al., model intercomparison studies include Energy Modeling Forum (EMF) 2014b; Rogelj et al., 2013a), there have been a number of multi-model 27 (Krey et al., 2014; Kriegler et al., 2014a), ADAM (Adaptation and projects exploring non-idealized implementation scenarios (Table 6.1). Mitigation Strategies: Supporting European Climate Policy) (Edenhofer This chapter relies heavily on those multi-model studies. et al., 2010), RECIPE (Report on Energy and Climate Policy in Europe) (Luderer et al., 2012a; Tavoni et al., 2012), and AMPERE (Assessment There are a number of ways that scenarios may deviate from the ideal- of Climate Change Mitigation Pathways and Evaluation of the Robust- ized implementation, but two are most prominent in the new litera- ness of Mitigation Cost Estimates) (Riahi et al., 2014). In addition ture. One set of scenarios includes those in which near-term mitigation to the large model intercomparison studies, a number of individual Table 6.1 | Multi-model studies exploring non-idealized international implementation Multi-Model Study Description EMF 22 (Clarke et al., 2009) Delayed participation (fragmented action) scenarios in which Organisation for Economic Co-operation and Development (OECD) countries begin mitigation immediately; Brazil, Russia, India, and China begin after 2030; remaining countries begin after 2050. Scenarios meet various 2100 concentration goals, with and without overshooting the concentration goal. EMF 27 (Blanford et al., 2014; Delayed and limited participation scenario with Annex I adopting 80 % emissions reductions until 2050, non-Annex I adopting a global Kriegler et al., 2014a) 50 % emissions reduction by 2050 after 2020, and resource exporting countries not undertaking emissions reductions. AMPERE (Kriegler et al., Two studies: AMPERE WP2 focused on delayed mitigation scenarios with the world following moderate 2014c; Riahi et al., 2014) mitgation until 2030, and adopting long-term concentration goals thereafter. AMPERE WP3 focused on delayed participation scenarios with EU27 or EU27 and China acting immediately and the remaining countries transitioning from moderate policies to a global carbon pricing regime (without mitigation goal) between 2030 and 2050. LIMITS (Kriegler et al., 2013b; Delayed mitgation scenarios with the world following two levels of moderate fragmented action through 2020 or 2030, and Tavoni et al., 2013) adopting two long-term concentration goals thereafter. Three different effort-sharing schemes are considered. RoSE (Luderer et al., 2014a) Delayed mitgation scenarios with the world following moderate fragmented action in the near term and adopting a long-term concentration goal after 2020 or 2030. Note: The Energy Modeling Forum (EMF) 27, AMPERE (Assessment of Climate Change Mitigation Pathways and Evaluation of the Robustness of Mitigation Cost Estimates), LIMITS (Low Climate Impact Scenarios and the Implications of Reguired Tight Emission Control Strategies), and RoSE (Roadmaps Towards Sustainable Energy Futures) studies also included scenarios of moderate fragmented action throughout the 21st century without the goal of meeting any specific long-term concentration. 421
Assessing Transformation Pathways Chapter 6 research papers and reports have explored this space since AR4, typi- transactions, information asymmetries, and market power influencing cally constrained to a single model (Richels et al., 2007; Calvin et al., decisions are not effectively represented. Maintaining a long-term, 2009a; Krey and Riahi, 2009; van Vliet et al., 2009; Riahi et al., 2012; integrated, and often global perspective involves tradeoffs in terms Luderer et al., 2013; Rogelj et al., 2013b). In many cases, these stud- of the detail at which key processes can be represented in integrated ies have simply assumed that particular technologies, such as CCS models. Hence, the models do not generally represent the behaviour or nuclear power, may not be available. In others, studies have put of certain important system dynamics, such as economic cycles or the 6 constraints on resource supplies, for example, the supply of bioenergy. operation of electric power systems important for the integration of In others, they have called for variations in cost and performance of solar and wind power, at the level of detail that would be afforded by different technologies. Many have also explored the implications of analyses that the focus exclusively on those dynamics. energy end-use improvements. Beyond these and other similarities, integrated modelling approaches can be very different, and these differences can have important impli- cations for the variation among scenarios that emerge from different 6.2 Tools of analysis models. The following paragraphs highlight a number of key differ- ences in model structure. To provide insight into the implications of these tradeoffs, potential implications for aggregate economic costs are provided as examples, when appropriate. 6.2.1 Overview of integrated modelling tools Economic coverage and interactions. Models differ in terms of the The long-term scenarios assessed in this chapter were generated degree of detail with which they represent the economic system and primarily by large-scale, integrated models that can project key char- the degree of interaction they represent across economic sectors. Full- acteristics of transformation pathways to mid-century and beyond. economy models (e. g., general equilibrium models) represent inter- These models represent many of the most relevant interactions among actions across all sectors of the economy, allowing them to explore important human systems (e. g., energy, agriculture, the economic and understand ripple effects from, for example, the imposition of system), and often represent important physical processes associated a mitigation policy, including impacts on overall economic growth. with climate change (e. g., the carbon cycle). Other approaches to Partial-economy models, on the other hand, take economic activ- explore transformation pathways include qualitative scenario methods ity as an input that is unresponsive to policies or other changes such and highly aggregated modelling tools, such as those used for cost- as those associated with improvements in technology. These models benefit analysis (see Box 6.1 on cost-benefit analysis, p. 394). These tend to focus more on detailed representations of key systems such other approaches provide a different level of quantitative information as the energy system. All else equal, aggregate economic costs would about transformation pathways than scenarios from large-scale inte- tend to be higher in full-economy models than in partial-economy grated models. models because full-economy models include feedbacks to the entire economy. On the other hand, full-economy models may include more All integrated models share some common traits. Most fundamentally, possibilities for substitution in sectors outside of those represented in integrated models are simplified, stylized, numerical approaches to partial-economy models, and this would tend to reduce aggregate eco- represent enormously complex physical and social systems. They take nomic costs. in a set of input assumptions and produce outputs such as energy system transitions, land-use transitions, economic effects of mitiga- Foresight. Perfect-foresight models (e. g., intertemporal optimization tion, and emissions trajectories. Important input assumptions include models) optimize over time, so that all future decisions are taken into population growth, baseline economic growth, resources, technologi- account in today’s decisions. In contrast, recursive-dynamic models cal change, and the mitigation policy environment. The models do not make decisions at each point in time based only on the information in structurally represent many social and political forces that can influ- that time period. In general, perfect-foresight models would be likely to ence the way the world evolves (e. g., shocks such as the oil crisis of allocate emissions reductions more efficiently over time than recursive- the 1970s). Instead, the implications of these forces enter the model dynamic models, which should lead to lower aggregate costs. through assumptions about, for example, economic growth and resource supplies. The models use economics as the basis for decision Representation of trade. Models differ in terms of how easy it is making. This may be implemented in a variety of ways, but it funda- for goods to flow across regions. On one end of the spectrum are mentally implies that the models tend toward the goal of minimizing models assuming goods are homogeneous and traded easily at one the aggregate economic costs of achieving mitigation outcomes, unless world price (Heckscher-Ohlin) or that there is one global producer they are specifically constrained to behave otherwise. In this sense, the (quasi-trade). On the other end of the spectrum are models assuming scenarios tend towards normative, economics-focused descriptions of a preference for domestic goods over imported goods (Armington) or the future. The models typically assume fully functioning markets and models without explicit trade across regions (e. g., models with import competitive market behavior, meaning that factors such as non-market supply functions). In general, greater flexibility to trade will result in 422
Chapter 6 Assessing Transformation Pathways lower-aggregate mitigation costs because the global economy is more 6.2.2 Overview of the scenario ensemble for flexible to undertake mitigation where it is least expensive. More gen- this assessment erally, many partial-equilibrium models include trade only in carbon permits and basic energy commodities. These models are not capable The synthesis in this chapter is based on a large set of new scenarios of exploring the full nature of carbon leakage that might emerge from produced since AR4. The number of models has increased and model mitigation policies, and particularly those associated with fragmented functionality has significantly improved since AR4, allowing for a international action. broader set of scenarios in the AR5 ensemble. The majority of these 6 scenarios were produced as part of multi-model comparisons. Most Model flexibility. The flexibility of models describes the degree model intercomparison studies produce publicly available databases to which they can change course. Model flexibility is not a single, that include many of the key outputs from the studies. Although crucial explicit choice for model structure. Instead, it is the result of a range for our understanding of transformation pathways, these intercompari- of choices that influence, for example, how easily capital can be reallo- son exercises are not the only source of information on transformation cated across sectors including the allowance for premature retirement pathways. A range of individual studies has been produced since AR4, of capital stock, how easily the economy is able to substitute across largely assessing transformation pathways in ways not addressed in energy technologies, whether fossil fuel and renewable resource con- the model intercomparison exercises. For the purposes of this assess- straints exist, and how easily the economy can extract resources. The ment, an open call was put forward for modellers to submit scenarios complexity of the different factors influencing model flexibility makes not included in the large model intercomparison databases. These clear delineations of which models are more or less flexible difficult. scenarios, along with those from many of the model intercomparison Evaluation and characterization of model flexibility is an area of cur- studies, have been collected in a database that is used extensively in rent research (see Kriegler et al., 2014b). Greater flexibility will tend to this chapter. A summary of the models and model intercomparison lower mitigation costs. exercises that generated the scenarios referenced in this chapter can be found in Annex II.10. Sectoral, regional, technology, and GHG detail. Models differ dra- matically in terms of the detail at which they represent key sectors and systems. These differences influence not only the way that the models 6.2.3 Uncertainty and the interpretation of operate, but also the information they can provide about transforma- large scenario ensembles tion pathways. Key choices include the number of regions, the degree of technological detail in each sector, which GHGs are represented and The interpretation of large ensembles of scenarios from different mod- how, whether land use is explicitly represented, and the sophistica- els, different studies, and different versions of individual models is a tion of the model of earth system process such as the carbon cycle. core component of the assessment of transformation pathways in this Some models include only CO2 emissions, many do not treat land-use chapter. Indeed, many of the tables and figures represent ranges of change (LUC) and associated emissions, and many do not have sub- results across all these dimensions. models of the carbon cycle necessary to calculate CO2 concentrations. In addition, although the scenarios in this chapter were generated There is an unavoidable ambiguity in interpreting ensemble results in from global models that allow for the implications of mitigation for the context of uncertainty. On the one hand, the scenarios assessed in international markets to be measured, regional models can provide this chapter do not represent a random sample that can be used for finer detail on the implications for a specific region’s economy and dis- formal uncertainty analysis. Each scenario was developed for a specific tributional effects. The effects of detail on aggregate mitigation costs purpose. Hence, the collection of scenarios included in this chapter does are ambiguous not necessarily comprise a set of ‘best guesses.’ In addition, many of these scenarios represent sensitivities, particularly along the dimensions Representation of technological change. Models can be catego- of future technology availability and the timing of international action rized into two groups with respect to technological change. On one on climate change, and are therefore highly correlated. Indeed, most of end of the spectrum, models with exogenous technological change the scenarios assessed in this chapter were generated as part of model take technology as an input that evolves independently of policy mea- intercomparison exercises that impose specific assumptions, often sures or investment decisions. These models provide no insight on regarding long-term policy approaches to mitigation, but also in some how policies may induce advancements in technology. On the other cases regarding fundamental drivers like technology, population growth, end of the spectrum, models with endogenous technological change and economic growth. In addition, some modelling groups have gener- (also known as induced technological change) allow for some por- ated substantially more scenarios than others, introducing a weighting tion of technological change to be influenced by deployment rates of scenarios that can be difficult to interpret. At the same time, however, or investments in research and development (R&D). Models featuring with the exception of pure sensitivity studies, the scenarios were gen- endogenous technological change are valuable for understanding how erated by experts making informed judgements about how key forces the pace of technological change might be influenced by mitigation might evolve in the future and how important systems interact. Hence, policies. although they are not explicitly representative of uncertainty, they do 423
Assessing Transformation Pathways Chapter 6 provide real and often clear insights about our lack of knowledge about are generally published. Whether certain circumstances are under- key forces that might shape the future (Fischedick et al., 2011; Krey and represented because they have been under-examined or because they Clarke, 2011). The synthesis in this chapter does not attempt to resolve have been examined and the scenarios failed is a crucial distinction, the ambiguity associated with ranges of scenarios, and instead focuses yet one that it is currently not possible to fully report. Model infeasibili- simply on articulating the most robust and valuable insights that can ties can bias results in important ways, for example, the costs of miti- be extracted given this ambiguity. However, wherever possible, scenario gation, because only those models producing scenarios can provide 6 samples are chosen in such a way as to reduce bias, and these choices estimated costs (Tavoni and Tol, 2010). are made clear in the discussion and figure legends. 6.2.4 Interpretation of model inability to 6.3 Climate stabilization: produce particular scenarios Concepts, costs and A question that is often raised about particular stabilization goals and implications for the macro transformation pathways is whether the goals or pathways are ‘fea- economy, sectors and sible’ (see Section 6.1). Integrated models can be helpful in informing technology portfolios, this question by providing information about key elements of transfor- mation pathways that might go into assessments of feasibility, such taking into account as rates of deployment of energy technologies, rates of reductions differences across regions in global and regional emissions, aggregate economic costs, finan- cial flows among regions, and links to other policy objectives such as energy security or energy prices. However, beyond cases where physi- 6.3.1 Baseline scenarios cal laws might be violated to achieve a particular scenario (for exam- ple, a 2100 carbon budget is exceeded prior to 2100 with no option for negative emissions), these integrated models cannot determine feasi- 6.3.1.1 Introduction to baseline scenarios bility in an absolute sense. Baseline scenarios are projections of GHG emissions and their key driv- This is an important consideration when encountering situations in ers as they might evolve in a future in which no explicit actions are which models are incapable of producing scenarios. Many models taken to reduce GHG emissions. Baseline scenarios play the important have been unable to achieve particularly aggressive concentration role of establishing the projected scale and composition of the future goals such as reaching 450 ppm CO2eq by 2100, particularly under energy, economic, and land-use systems as a reference point for mea- challenging technological or policy constraints. In some cases, this suring the extent and nature of required mitigation for a given climate may be due to the violation of real physical laws, the most common of goal. Accordingly, the resulting estimates of mitigation effort and costs which is when the cumulative carbon budget associated with meeting in a particular mitigation scenario are always conditional upon the a long-term goal is exceeded without options to remove carbon from associated baseline. the atmosphere. Frequently, however, instances of model infeasibility arise from pushing models beyond the boundaries of what they were Although the range of emissions pathways across baseline scenarios in built to explore, for example, rates of change in the energy system that the literature is broad, it may not represent the full potential range of exceed what the model can represent, or carbon prices sufficiently possibilities. There has been comparatively little research formally con- high that they conflict with the underlying computational structure. structing or eliciting subjective probabilities for comprehensive ranges Indeed, in many cases, one model may be able to produce scenarios of the key drivers of baseline emissions in a country-specific context, while another will not, and model improvements over time may result and this remains an important research need for scenario develop- in feasible scenarios that previously were infeasible. Hence, although ment. As discussed in Section 6.2, although the range of assumptions these model infeasibilities cannot generally be taken as an indicator of used in the literature conveys some information regarding modellers’ feasibility in an absolute sense, they are nonetheless valuable indica- expectations about how key drivers might evolve and the associated tors of the challenge associated with achieving particular scenarios. implications, several important factors limit its interpretation as a true For this reason, whenever possible, this chapter highlights those situa- uncertainty range. An important distinction between scenarios in this tions where models were unable to produce scenarios. regard is between those that are based on modellers’ ‘default’ assump- tions and those that are harmonized across models within specific Unfortunately, this type of result can be difficult to fully represent in studies. The former can be considered a better, although still imperfect, an assessment because, outside of model intercomparison studies representation of modellers’ expectations about the future, while, as is intended explicitly to identify these circumstances, only scenarios that discussed below, the latter consider specific alternative views that in could actually be produced (as opposed that could not be produced) some cases span a larger range of possible outcomes. 424
You can also read